Hospital-acquired acute kidney injury: An analysis of nadir-to-peak serum creatinine increments stratified by baseline estimated GFR

Background. Current working serum creatinine-based definitions of acute kidney injury (AKI) do not take into consideration the baseline level of kidney function, as defined by the estimated glomerular filtration rate (eGFR). The objective of this study was to explore whether the baseline eGFR should be taken into account when forming a definition of AKI.;Study design. Single-center retrospective cohort study.;Setting and participants. 29,645 adult subjects hospitalized at least once at an acute care hospital between 2000 and 2007.;Predictor. Hospital-acquired AKI was defined by calculating the difference between the peak and nadir serum creatinine. The nadir serum creatinine was ascertained in the first 3 days of the hospitalization, and was used to calculate the baseline eGFR, using the 4 variable MDRD Study equation.;Outcomes. In-hospital mortality, and among survivors, discharge disposition (to a facility vs. home).;Results. Different thresholds of serum creatinine were found to be independently associated with increased mortality according to different eGFR strata. Indeed, a nadir-to-peak serum creatinine of ≥ 0.2, ≥ 0.3 and ≥ 0.5 mg/dl were found to be associated with increased mortality in patients with baseline eGFR ≥ 60 (odds ratio [OR] 1.67; 95% confidence interval [CI] 1.13, 2.47), 30-59 (OR 2.69; 95% CI 1.82, 3.97), and < 30 (OR 2.15; 95% CI 1.02, 4.51) ml/min/1.73 m2, respectively, after multivariable adjustment . There was a significant interaction of the nadir-to-peak serum creatinine and baseline eGFR strata for the outcome of in-hospital mortality (P < 0.001), and among hospital survivors, with hospital discharge to facility (P < 0.001). Sensivity analyses displayed similar results, with a significant interaction between eGFR strata and relative increases in serum creatinine, as well as with absolute and relative decreases in eGFR (P < 0.001 for all analyses).;Limitations. Single center study; residual confounders; untimed definition of nadir-to-peak serum creatinine.;Conclusions. This study suggests that future development of serum creatinine-based definitions of AKI should take into consideration baseline kidney function.